Central-Intel is a personal AI relay platform. It connects you to AI models from multiple providers (OpenAI, Anthropic, Google, and others), runs local models on your device via llama.cpp, and enriches every conversation with your personal knowledge base, memory, and integrations — all while keeping your data private and encrypted.
How do I get started?
Download the app for your platform (desktop, Android, or iOS). The Free tier works immediately with local models — no account needed. To use cloud AI providers on the Free tier, add your own API keys in Settings. For relay features (device sync, memory sync, desktop tunnel), sign up for a Basic or Pro plan.
What devices are supported?
Central-Intel is available on macOS, Windows, Linux (desktop), Android, and iOS. All platforms support local models, cloud providers, profiles, RAG, and memory. Device sync and memory sync require a Basic or Pro plan.
Do I need an account?
No! The Free tier works entirely offline with local models, or with your own API keys called directly from your device. An account is only needed for relay features (Basic/Pro), device sync, and memory sync.
Pricing & Billing
What are the pricing tiers?
We offer three tiers: Free ($0, local models + direct BYOK), Basic (pay-as-you-go credits for relay access), and Pro ($16/mo or $156/year — everything in Basic plus BYOK through relay, Pro pricing on credits, and priority support). See our full pricing page for details.
How do credits work on the Basic plan?
Credits are prepaid balance that covers the cost of AI requests routed through our relay. When you send a message using a cloud provider through the relay, credits are deducted based on the model used and token count. You can top up credits anytime from your dashboard. You only pay for what you use.
What's the difference between Free and Basic?
Free lets you use your own API keys directly (no relay involved — the app calls the provider from your device). Basic gives you relay access with managed credits, device sync, memory sync, and desktop tunnel. The relay lets your requests be enriched with context across devices.
Can I use my own API keys on the Basic plan?
BYOK (bring your own keys) through the relay is a Pro feature. On Basic, you purchase credits which cover both the AI provider costs and relay infrastructure. On Free, you can use your own keys directly (no relay) — the app calls the provider from your device.
Why can't Basic users use their own keys through the relay?
Running the relay costs us money — servers, bandwidth, and infrastructure. Pro subscribers cover that with their monthly fee, so they can bring their own keys through the relay without needing credits. Basic users cover relay costs through credits.
Can I switch between tiers?
Yes. You can upgrade to Pro at any time from your dashboard. If you cancel Pro, you'll retain Pro benefits until the end of your billing period, then revert to Basic. Your credit balance carries over between tiers.
How do I cancel my Pro subscription?
You can cancel from your dashboard under account settings. Cancellation takes effect at the end of your current billing period — you won't be charged again, and you'll keep Pro benefits until then.
Yes. Your conversations, documents, memory items, and knowledge base live on your device. If you enable cross-device memory sync, data is end-to-end encrypted (AES-256-GCM) — our server cannot read it. We never sell, share, or use your data for any purpose other than providing the service. See our Privacy Policy for full details.
What data do you collect?
Only what's needed to make the app work: your email address (for sign-in), device name (for device management), billing metadata (model used, token counts, cost — never message content), and a session cookie. If you opt in, anonymous usage analytics (no PII). See our Privacy Policy for the complete list.
How is my data encrypted?
All server communication uses HTTPS/TLS. API keys are stored in your OS keychain (iOS Keychain, Android Keystore, or desktop encrypted storage). Memory sync uses AES-256-GCM with a vault key that only your devices possess — our server handles encrypted blobs it cannot decrypt. Device identity uses Ed25519 signing and X25519 key agreement.
Does Central-Intel train AI models on my data?
No. Central-Intel never uses your data for AI training. We are a relay — we route your requests to the AI provider you choose. When you use third-party AI providers, their own policies apply — we recommend reviewing them. See the provider list in our Privacy Policy.
What about my health data?
Health data (via Android Health Connect) is pulled at the time of your request only — it's used to give your AI assistant context for health-related questions. It is never stored on our servers, never written to any database, and never transmitted outside of the AI request you initiated. You can revoke health permissions anytime in the app.
Local Models
What local models are supported?
Central-Intel supports Llama (Meta), Gemma (Google), and Qwen (Alibaba) models via llama.cpp. Models run entirely on your device — prompts never leave your machine. You can also connect to Ollama instances on your local network.
How do I run models locally?
On first launch, the app will offer to download a local model. Choose a model size that fits your device's RAM. Once downloaded, local models are available immediately — select them from the model picker in any chat. No API key or account needed.
Do local models require internet?
Only for the initial download. Once a model is on your device, it runs completely offline. You can use Central-Intel with local models in airplane mode, on a plane, or anywhere without connectivity.
Integrations
✓Health Connect — Steps, heart rate, sleep, nutrition, exercise, and more (Android)
✓On-Device Calendar — Local calendar events for schedule-aware context